Chromatography Forum

We manufacture cosmetic lotions containing methylparaben and propylparaben. We are getting ready to validate an HPLC method for paraben analysis. I would like to know if there is any recommendation for the specification on accuracy. Currently during preliminary testing, I am getting recoveries in the range of 94% to 103%. I have tried different columns, different extraction methods and cannot improve the recoveries. Please let me know if there is a standard industry guidance on the Specification such as +/- 4.0% or 10.0% etc for preservative accuracy. Thanks a million for you reply.

US: we also do parabens on consumer lotions, but our precision/accuracy/recovery is much better. Are you incorporating an acid such acetic acid in your mobile phase? We dissolve the samples in methanol and filter, then HPLC.

Here's what we obtained:
"Mean recoveries from regular scent placebo product spiked with three different known levels of each paraben (bracketing current use levels) were 99.0% with a standard deviation of 2.0 for methyl paraben and 100.0 with a standard deviation of 2.1 for propyl paraben with column choice #2 and 100.0% with a standard deviation of 2.1 for methyl paraben and 100.6 with a standard deviation of 1.9 for propyl paraben with column choice # 3".

Can you give more details on Column#2 and #3. we use Acetonitrile for extractions. The mobile phase is 50:50 0.1% H3PO4 and Acetonitrile and column we use is C18, micro-Bondapak from Waters.

Column 2 is a 2.1mm id type A RP-18 silica, 10 cm long with a shorter 3 cm cartridge of same as guard column. Column 3 is a modern "intrinsically base-deactivated" RP-18 sold by major vendors such as Thermo Keystone. The question is: are your peak shapes OK, and if so, is placebo product flat in the regions of interest? These parabens can also be determined by GC, but I'd trimethylsilyl derivatize first for that.